The Case Against a Special Regime for Intragroup Transactions

Corporate groups with minority shareholders in one or more subsidiaries are common around the world, despite the risks such arrangements pose to those shareholders. Shaping a firm as a web of formally independent, minority-co-owned legal entities facilitates controllers’ diversion of corporate wealth (tunnelling) via intragroup transactions and other non-transactional techniques. While many jurisdictions leave the regulation of intragroup transactions to ordinary remedies against self-dealing, others (mostly in Europe) establish a special regime centred on a relaxation of directors’ fiduciary duties. Under this special regime, subsidiary directors are not liable if they make disadvantageous decisions that are beneficial to other entities within their group, provided that proper compensation is granted (or, according to some proposals, may reasonably be expected to be granted) to the subsidiary. This article conducts a qualitative cost-benefit analysis of the special regime, focusing on the European Model Companies Act’s rules on intragroup transactions. We concede that such rules have the advantage of reducing contracting costs and enhancing managerial flexibility within the corporate group, relative to systems governed by ordinary corporate law rules against unfair self-dealing. However, we also show that those benefits can be expected to be very limited. Furthermore, we show that this special regime substantially reduces minority shareholder protection against tunnelling, by making it much harder for minority shareholders to recover damages from controllers’ unfair self-dealing. Overall, our analysis suggests that, for corporate groups with minority shareholders at the subsidiary level, this regime should be implemented as an opt-in arrangement, if at all. Even in that form, it should be adopted together with adequate protections for shareholders dissenting from the midstream resolution to opt into the regime.


Introduction
Virtually every major firm is organised as a group of companies rather than as one legal entity. 1 Businesses are shaped into a web of as many companies as is costeffective from the perspective of their (or more precisely their owners' or controllers') financing, governance, tax, regulatory and operational needs. In some of these business groups around the world, minority investors own equity stakes in one or more subsidiaries. 2 We term such groups as minority-co-owned groups (MCOGs) and keep our focus on this subset of groups throughout this article.
MCOGs are widely acknowledged to facilitate tunnelling, meaning controllers' appropriation of corporate wealth to the detriment of (non-controlling) shareholders. 3 Structuring the business as a network of formally independent entities that, being part of the same economic organisation, routinely transact with each other is a tremendously effective way of (i) multiplying the opportunities for value diversion and (ii), as IGTs become routine, making such value diversion harder to detect. 4 Furthermore, choosing the MCOG structure allows controllers to extract private benefits through non-transactional techniques that, by not involving formal exchange across group affiliates, are very difficult to police. 5 At the same time, however, MCOGs can be valuable organisational tools. 6 For instance, publicly traded MCOGs permit a larger and more fine-tuned use of performance-based compensation, in the form of stock options or similar arrangements. 1 See, e.g., Dau et al. (2021), at p 161 ('Business groups […] are not only prevalent across much of the globe but, in many countries and regions, are the primary form of business organization'); Hopt (2015), at p 603 ('Groups of companies rather than single independent companies are the modern reality of the corporation'). For a comprehensive study of business groups in western countries, also containing empirical data, see Colpan and Hikino (2018). 2 See, e.g., Dau et al. (2020), at p 3 (observing that '[i]n many economies, most large listed companies came to belong to one of a handful of business groups'). This is rather the exception than the norm in the U.S., where subsidiaries are most often wholly owned. See, e.g., Squire (2011), at p 611. For two examples of U.S. minority-co-owned groups, one involving Coca-Cola, see Atanasov et al. (2011), at pp 29-33. 3 For a useful taxonomy of tunnelling techniques see generally Atanasov et al. (2011), at pp 5-9. For empirical evidence of high tunnelling in Korean and Indian groups see, respectively, Baek et al. (2006); Bertrand et al. (2002). 4 See, e.g., Hopt (2015), at p 619; Enriques (2018), at p 508. 5 E.g., controllers may prevent the minority co-owned affiliate from engaging in a profitable project that may harm other group affiliates (those where the controller has a larger equity stake) or they may force the affiliate into the development of a loss-generating project that generates positive externalities for other group affiliates (again, those where the controller has a higher equity interest). In both cases, value is transferred from the affiliate's minority shareholders to the controller without resorting to intra-group transactions. Quite intuitively, the absence of any visible transaction makes many corporate law remedies against unfair self-dealing not applicable to this tunnelling technique. Non-transactional tunnelling received much less attention than transactional tunnelling in the scholarly literature. There are, however, exceptions: see Enriques (2015), at pp 10-11; Dammann (2008), at pp 693-694. A thorough analysis of non-transactional tunnelling is provided by Bebchuk and Hamdani (2017), at pp 21-39 (discussing 'indirect tunnelling' as a form of value diversion typically occurring when a controlling shareholder owns other businesses in related industries). 6 See Enriques and Gilotta (2024; forthcoming) for a more detailed analysis of justifications for MCOGs and their policy implications.
Managers of a listed subsidiary will be better incentivised to perform if they are paid in the subsidiary's stock. Furthermore, listing the shares of one or more subsidiaries may provide investors with more granular information about the group's operations and more credibly bind controllers to enhanced transparency at the business unit level.
MCOGs are the subject of a sharp divide across national corporate laws. Some jurisdictions leave the regulation of relationships between members of the same group to the common rules of corporate law, thus subjecting MCOGs to ordinary rules against directors' and controlling shareholders' self-dealing. Other jurisdictions, on the contrary, establish a special regime, usually centred on the relaxation of directors' fiduciary duties with regard to such relationships, possibly offsetting this more lenient regime by introducing other structural (that is, not focused on the single transaction or behaviour) requirements, such as exit rights at the time when a company becomes part of a group. 7 Hereinafter, for brevity, we dub this special regime 'group law'.
In this article, we conduct a qualitative cost-benefit analysis of group law. 8 Our goal is to shed new light on the following key question in corporate law and governance: do efficiency considerations justify the relaxation of fiduciary duties and, where present, other constraints on self-dealing, in firms organised as groups? Answering this question is important from a policy perspective, as it would help policymakers design better corporate law rules for an organisational form-namely the MCOG-that is widely diffused worldwide. 9 The primary goal of special rules on groups is ultimately, to grant the controller (the parent company or its ultimate controlling shareholder, if there is one) greater leeway in managing the group. The underlying justification is that freeing group controllers from constraints on their ability to manage the group as though it was a single entity reduces the costs of managing the group and enhances the parent's ability to maximise group value, to the benefit not only of the controller but also of society as a whole and, possibly, minority shareholders themselves. 10 7 See Sect. 2.1. 8 A quantitative cost-benefit analysis-i.e., an empirical investigation aimed at measuring the costs and the benefits of group law in any particular jurisdiction where it applies (see Sect. 2.1)-may possibly provide a more precise answer to the question whether group law is a desirable piece of legislation and therefore stronger guidance to policymakers. Unfortunately, the present authors would have no comparative advantage in conducting such an investigation. But we find it telling that, to the best of our knowledge, no such empirical study has been conducted so far. 9 See, e.g., Masulis et al. (2011), at pp 3569-3570 (showing that in many countries a significant percentage of listed firms belongs to larger business groups). 10 To be sure, whether group law benefits society as a whole rather than merely the controller will depend on whether such benefits will be greater than any costs arising from group law. See infra nn. 19-20 and accompanying text. Whether group law also brings benefits to minority shareholders of MCOGs chiefly depends on whether the surplus arising from lower group management costs will be shared with them to an extent that allows for the compensation of harm deriving, inter alia, from the greater amount of tunnelling that the relaxation of corporate law constraints against self-dealing will enable.
The most significant obstacle to smoother group management is usually identified in directors' fiduciary duties, namely (and whatever the label in individual jurisdictions) the duty of loyalty. Group law proponents view standard fiduciary duties as a hindrance to efficient group management and hence to controllers' ability to maximise group value. According to them, such duties impose excessive constraints on intragroup exchange and group members' investment policies, thus hindering a number of value-creating actions. Fiduciary duties, as the argument goes, imply an assessment of fairness (and hence, presumptively, value creation) focused on each individual transaction, without allowing for a broader perspective that considers intragroup exchange as non-episodic and therefore capable of offsetting the harm suffered from an individual transaction with the benefits gleaned from another. The narrow focus of ordinary corporate law fiduciary duties may deliver 'false positives', i.e., overall fair (and value-creating) intragroup exchange that never materialises or is mistakenly judged to be unfair. If the risk exists of fair and efficient IGTs being screened out by internal decision-making bodies or struck down by courts, grouplevel value maximisation is not achieved.
A common element of group law across the jurisdictions that have adopted it is that controllers are granted the licence to force the affiliate to make disadvantageous decisions for itself, giving controllers flexibility in managing the group: they may in fact allocate resources and business opportunities free of fair-price constraints (i.e., as though the group were a single multi-divisional firm, where no such constraints apply to inter-divisional exchange). Yet, this greater managerial freedom is usually subject to one important condition: any harm inflicted on the subsidiary as a consequence of unfair transactions must be, within a certain timeframe, fully compensated. 11 In an even more enabling formulation of this regime, what is necessary and sufficient is that, at the time the harm is inflicted, directors can reasonably assume such compensation to materialise within a reasonable time. 12 In short, in the view of its supporters, a special enabling regime on intragroup transactions is an efficient policy that is apt to maximise the value of firms organised as groups, allowing shareholders to capture the greatest possible benefits from choosing this organisational structure.
Our analysis supports a more sceptical view. We show that, as far as MCOGs are concerned, the benefits of adopting this regime are limited and the costs substantial. To reach our conclusion, we focus on the rules on groups contained in the European Model Companies Act (EMCA), a model law drafted by a group of European academics. 13 Doing so allows us to avoid the impracticality of referring to a multitude of slightly different national legal regimes and, at the same time, to (at least partially) deflect any potential criticism issued for building our own strawman regime on groups. The EMCA's rules build upon national solutions and scholarly contributions to the debate on how to regulate corporate groups 14 and are centred on the standard according to which directors of a subsidiary may legitimately adopt decisions harmful for their company if (1) the decision is beneficial to the group as a whole and (2) at the time the harmful decision is taken, directors may reasonably assume that the harm will be offset by a benefit within a reasonable time (hereafter, the EMCA standard). 15 We should acknowledge at the very start that focusing on an open-ended standard that no jurisdiction has adopted verbatim prompts us to engage in tentative speculations about how to interpret it. 16 However, the EMCA standard is quite similar to some national group laws. 17 Hence, it is not unreasonable to assume that courts' interpretations of the former would not radically depart from the most common and widely accepted interpretations of the latter.
Our critique debunks the idea that the EMCA standard brings about significant net benefits, showing that the advantages usually associated with it are more limited than its proponents contend and that the costs are larger than usually believed.
First, we show that any type of group-value-maximising action that can be implemented under the EMCA standard can in principle also be implemented under ordinary director duties. Most notably, we show that director duties, including the core duty to act in the company's best interest, do not prevent group company directors from undertaking actions that create additional net value at the group level, at the expense of the affiliate company. Indeed, what ordinary principles require in this case is that the company receives adequate compensation (or a legally enforceable promise thereof)-a negotiating outcome that, transaction costs aside, can always be reached, given that, by hypothesis, the harmful action for which compensation is required creates overall more wealth than it destroys. Ordinary director duties thus offer more flexibility than many scholars tend to think, which means that the EMCA standard offers less additional flexibility than is commonly believed.
Second, while we acknowledge that the EMCA standard allows for the reduction of the transaction costs of actions increasing group value, we show that the magnitude of this cost reduction can be expected not to be significant.
Third, we show that the EMCA standard is unable to provide a better assessment of fairness in intragroup exchange than the ordinary corporate law rules on selfdealing. The principle may help to reduce the number of false positives (transactions that do not divert value but are mistakenly considered to be value-diverting), but it also increases the number of false negatives (transactions that divert value but are mistakenly considered as not value-diverting). 14 See especially Conac (2013); Conac (2016);Teichmann (2013);Teichmann (2016). 15 See infra n. 103 and accompanying text. 16 We should also acknowledge that the EMCA standard goes furthest in the direction of flexibility in the management of corporate groups. Hence, some of our critiques will not apply, or apply less forcefully, to less extreme solutions, such as those found in German corporate law (for a description, see Sect. 2.1). 17 Especially to those in France and Italy. See Sect. 2.1 for an account of such group laws.
In fact, in the face of its limited benefits, the EMCA standard raises significant concerns regarding minority shareholder protection against tunnelling. Minority shareholders would have a hard time persuading a court that directors are liable for breach of their fiduciary duties in the case of an IGT. Under the reasonable assumption that the challenged transactions will be a subset of those diverting value from subsidiaries because detection will be difficult and costly for minority shareholders, the EMCA standard offers controllers the opportunity to single out ex post whatever benefit an affiliate may have obtained from its participation in the group and 'spend' it to escape liability for those value-diverting actions minority shareholders have detected and sued over. Worse still, minority shareholders may also encounter undue difficulties in recovering damages with respect to harmful transactions for which ex post compensation is lacking, since in this case directors may still avoid liability by proving that it was nonetheless reasonable, ex ante, to expect compensation. 18 The higher the chances of escaping liability for unfair self-dealing, the more controllers will be inclined to engage in tunnelling. If minority shareholders correctly discount the negative effects of tunnelling on the share price, they may still get a fair deal. But a lax regime would also have negative effects more broadly, in the form of higher agency costs 19 and a higher cost of capital across the board, which may lead to fewer listings. 20 Overall, our analysis suggests that EMCA-style special regimes for intragroup transactions are dysfunctional as far as MCOGs are concerned. They bring about some benefits (e.g., lower group management costs and increased financing opportunities) but these benefits appear overall modest compared to the costs they entail (specifically a higher tunnelling risk). 21 As the article proceeds, Sect. 2 focuses our discussion on group law in continental Europe, where legal scholars have traditionally supported the idea of establishing a special corporate law regime for groups. We briefly recall the main features of national group laws in Germany, France and Italy, and summarise the positions of the various expert groups which have supported the idea of establishing a special EU regime for groups. Thereafter, we provide an account of EU policymakers' 18 To be clear, this is a feature of the EMCA standard that is not observed in any individual jurisdiction, perhaps with the exception of France, given the vague contours of the Rozenblum doctrine. See infra Sect. 2.1. 19 See generally Jensen and Meckling (1976), at p 308 (identifying agency costs as the sum of bonding costs, monitoring costs and the 'residual loss' from agents' suboptimal decisions). 20 See Djankov et al. (2008), at p 431; Pacces (2011), at p 192. 21 Only in exceptional circumstances may the benefits outweigh the costs. One such circumstance is where non-legal constraints already curb controllers' tunnelling in an effective way, so that legal remedies against that behaviour essentially do not matter. Where that is the case, basic legal constraints against unfair self-dealing, such as fiduciary duties, may be relaxed without major concerns. Another case in point might be where the entire economic system has been hit by large-scale unforeseen events, such as a pandemic: in this scenario most companies are in 'survival mode' and the primary goal for any policymaker is to keep afloat as many of them as possible. In this case, it might be preferable for lean decision-making to trump minority shareholder protection. Cf. Enriques (2020), at p 266. Special regimes for groups such as the EMCA standard might prove useful in these truly exceptional circumstances, because to some extent they do increase managerial flexibility in firms organised as a web of different legal entities.
attempts to establish such a special regime and describe the EMCA rules on groups. Section 3, the core of this article, delves into the practical implications of group law, as epitomised by the EMCA standard: on the one hand, it slightly increases the degree of flexibility in group management by reducing the costs associated with the implementation of group-value-maximising strategies; on the other, it greatly facilitates tunnelling. Section 4 concludes with a summary of the article's main results and suggests that, if at all a special, lenient standard for IGTs and, more broadly, for decisions made in the interest of (other entities of) the group is justified, it should only operate as an opt-in regime. In addition, any midstream transition to it should be accompanied by adequate safeguards for minority shareholders.

Towards an Enabling Law for European Corporate Groups?
Corporate law generally addresses tunnelling risks via fiduciary duties and procedural rules aimed at sterilising conflicts of interest in corporate decision-making. 22 Fiduciary duties require directors to act in the company's best interest. Absent special rules, any deviation from this broad standard, such as prioritising the interest of the parent company over that of the subsidiary, entitles the subsidiary (and, in most jurisdictions, the minority shareholders on its behalf) to obtain damages from its directors (and sometimes the parent company). Procedural rules regulate the company's internal decision-making processes with a view to minimising the risk of conflicted transactions' terms being unfair to the company. This is achieved through various techniques, such as ex post judicial review of the transaction's procedural fairness, enhanced disclosure requirements and transaction approval by disinterested decision-makers (such as independent directors or unconflicted shareholders). 23 Group law supporters argue that groups-including MCOGs-should be governed by different rules. 24 In their view, directors of a group subsidiary should be allowed to take the 'interest of the group' into account when managing the subsidiary and, subject to certain conditions, prioritise that interest over the interest of the subsidiary (i.e., taking decisions that are disadvantageous for the subsidiary but beneficial for the group).
As we show in this section, this idea is especially popular in continental Europe. Indeed, group law rules allowing for such prioritisation are a long-standing reality there. In addition, they have long been on the agenda of EU policymakers and have been called for by a number of expert groups, including the EMCA drafters. 22 See Enriques (2015), at pp 1521. 23 For an overview see Enriques et al. (2017), at pp 147-165. 24 See infra Sect. 2.2. We are aware that not all commentators advocating the introduction of a special regime for groups argue that wholly owned groups and MCOGs should be subject to the same set of special rules. To the contrary, many argue that some form of additional protection should be put in place where there are minority shareholders at the subsidiary level. See, e.g., the position of the Forum Europaeum on Company Groups recalled infra n. 76 and accompanying text. However, the fact remains that in many national group laws, as well as in the policy proposals of most group law supporters, deviations from ordinary corporate law rules against unfair self-dealing are established also for MCOGs.

The National Group Laws of Germany, France and Italy
Germany, France and Italy have long had special rules on groups in their national corporate laws. Our goal here is not to provide a comprehensive account of such special group laws and the issues they raise (let alone to offer a complete picture of how, more generally, corporate law in those countries regulates corporate groups and deals with the protection of minority shareholders therein), but to provide readers who are unfamiliar with these peculiar regimes with some background information on their core features. This will also help to set the stage for our discussion of the EMCA rules below. 25 German group law (Konzernrecht) represents the oldest and most influential 26 regime for corporate groups. It distinguishes between contractual groups and de facto groups. Contractual groups are groups based upon an explicit agreement between the parent and the subsidiary, according to which the latter chooses to subject itself to the parent's instructions. 27 In de facto groups this agreement is absent, with the parent's influence over the subsidiary stemming, usually, from the controlling equity stake that the former owns in the latter. 28 German group law allows directors of a subsidiary in a de facto group to adopt decisions disadvantageous to one company that benefit the whole group (that is, decisions that are contrary to the subsidiary's interest), 29 provided, though, that the subsidiary receives full compensation for the damages suffered. 30 Violation of these rules renders the parent, its directors and the directors of the subsidiary liable. 31 25 See Sect. 2.3. 26 Over time, German group law has inspired corporate lawmakers in several European countries, such as Portugal, Czech Republic, Slovenia, and Hungary. See Hopt (2015), at p 612. 27 See AktG, § 291, para. 1, first sentence. 28 See most recently Hopt (2015), at p 613. 29 Ibid. 30 See AktG, § 311. Directors of the subsidiary are also required to prepare a yearly report (known as the 'dependency report') containing a detailed description of the transactions between the subsidiary and other group affiliates (see AktG, § 312). The report remains confidential (i.e., shareholders and the larger public have no access to it) [Hopt (2015), at p 617] but it must be audited by the firm in charge of auditing the company's financial statements (see AktG, § 313). In addition, the subsidiary's supervisory board has to examine the report and report on it to the general meeting (see AktG, 314). If issues emerge as a result of these controls, shareholders have the right to ask the court for a special investigation (see AktG, § 315). The dependency report's main function is to provide evidence that any harm inflicted on the subsidiary as a consequence of disadvantageous transactions in the interest of the group has been properly compensated. See, e.g., Informal Company Law Expert Group (ICLEG) (2016), at p 6 (documenting that the report 'quantifies the disadvantages inflicted on the subsidiary and the compensations granted to the subsidiary to balance those disadvantages'); Hommelhoff (2001), at pp 67-69 (where also a discussion of the efficacy of the report in protecting minority shareholders). 31 See AktG, § § 317 and 318. Controllers of contractual groups, instead, are granted much greater room for maneuvering: within such groups, the subsidiary may adopt whatever disadvantageous decision that benefits the group, so long as the corporation's existence is not threatened: see, e.g., Tarde (2018), at pp 162-171.
French group law is judge-made and relies on the Rozenblum doctrine, first articulated by the French Cour de Cassation in 1985. 32 The Rozenblum doctrine creates a safe harbour (or 'group defence') against both criminal (abus the biens sociaux) 33 and civil liability 34 for subsidiaries' directors in 'integrated' company groups, namely groups 'characterized by capital links between the companies and by strong, effective business integration among the companies within the group'. 35 According to this doctrine, directors of a subsidiary who adopt a harmful decision that benefits the group are not held liable if the harm so inflicted does not threaten the subsidiary's solvency and if the subsidiary receives an economic quid pro quo that offsets that disadvantage. The economic quid pro quo, which should not be 'grossly inadequate,' 36 may however be of a non-monetary nature 37 and need not be received within a rigid and pre-established timeframe. 38 Italian group law establishes a standard according to which the parent is liable towards creditors and minority shareholders of group affiliates for abusing its influence over the subsidiaries. 39 The relevant provision allows minority shareholders and creditors to sue the parent for the damages resulting from such behaviour. 40 However, the same provision states that defendants are not liable if the damage inflicted on the subsidiary either is lacking in the light of the overall impact on the subsidiary of the (parent's) activity of group direction and coordination or has been fully offset via other transactions, including ad hoc ones. 41 32 Cour de Cassation, Chambre criminelle, Feb. 4, 1985, no. 84-91.581, in Revue des Sociétés 648 (1985. 33 See Art. L. 242-6, French C. Com. 34 See Informal Company Law Expert Group (ICLEG) (2016), at p 23. 35 See Antunes et al. (2011), at pp 62-63. 36 See Conac (2020), at p 92. 37 Conac (2013), at p 218. 38 See, e.g., European Company Law Experts (ECLE) (2016), at p 35. 39 See Art. 2497, para. 1, Civil Code. The provision is silent on whether the parent is also liable towards the subsidiary. Commentators hold diverging views on the issue: compare Abbadessa (2008), at pp 280-285 (arguing that the subsidiary is not entitled to sue the parent) with Cariello (2003), at p 339 (holding the opposite view). 40 Art. 2497, para. 1 and para. 3. Minority shareholders and creditors may also claim damages from anybody who took part in the damaging action and from anybody who knowingly took advantage of such action (in the latter case, however, minority shareholders may claim only an amount equal to what the defendant actually gained). It must also be noted that according to Art. 2497, para. 3, minority shareholders and creditors are entitled to sue the parent only if their company fails to indemnify them. This has awkward consequences: see infra n. 159. 41 See Art. 2497, para, 1, Civil Code ('Non vi è responsabilità se il danno risulta mancante alla luce del risultato complessivo dell'attività di direzione e coordinamento ovvero integralmente eliminato anche a seguito di operazioni a ciò dirette'). Similarly, the provision criminalising duty of loyalty breaches in specific circumstances provides that no criminal liability arises from IGTs if, the gain for the parent or sister company [is] offset by the advantages, whether realized or reasonably expected to be realized ['fondatamente prevedibili'], deriving from the correlation with other entities or from being part of the group' (Art. 2634 Civil Code). Offsetting advantages have legal relevance also in bankruptcy. Art. 284, para. 4, of the new bankruptcy code (Legislative Decree 12 January 2019, n. 14) establishes that the benefits that a group restructuring plan is expected to provide to creditors can be assessed also taking past or reason-Similarly to French group law, no timeframe is identified for these offsetting benefits to materialise and, at least according to the view of many Italian legal scholars, the advantage itself may be of a non-monetary nature, need not even be rigidly proportionate to the harm, 42 and, even for director liability purposes, may just be reasonably expected to be realised. 43 Group laws in Germany, France and Italy have one element in common: they all allow directors of a controlled company to adopt decisions that, in pursuing the interest of (other entities of) the group, impose an immediate harm to their company, provided that proper compensation is received for such inflicted harm.
What constitutes proper compensation, in turn, varies across the three jurisdictions. German law seemingly adopts the most stringent approach. Under Konzernrecht, the subsidiary must receive full compensation by the end of the financial year. Failing that, a determination has to be made by the same date as to the timing and the substance of the compensation for the harm suffered by the subsidiary. 44 If the disadvantage is quantifiable and reportable (meaning that it shows on the subsidiary's balance sheet), compensation must be of an equally reportable nature. 45 In any event, compensation must be appraisable. 46 Non-quantifiable advantages do not qualify as proper compensation. 47 The French and Italian standards are laxer. 48 According to both, no strict deadline for compensation is established. 49 Furthermore, the offsetting benefit need not necessarily be exactly quantifiable and/or reportable. 50 The German approach thus 42 See, e.g., Montalenti (1995), at p 731. But see contra Denozza (2000), pp 330, 338 (arguing that offsetting advantages must be, among other things, exactly quantifiable). Although the two works cited in this note predate the 2003 reform that first introduced a statutory law of groups in Italy, their arguments are still valid to ground alternative interpretations of the post-reform regime described in the text. 43 See Montalenti (2018), at p 895, using the same language as the criminal and bankruptcy law provisions referred to supra at n. 41. In order to offset potential risks for minority shareholders, Italian law establishes specific reporting and disclosure rules for decisions adopted under the parent's influence (decisions adopted under the parent's influence must be analytically motivated and accompanied by a precise indication of the reasons and the interests that affected the decision. Furthermore, they must be reported in the company's financial statements. See Art. 2497-ter Civil Code) and provides shareholders with an exit right when the company enters into the group (see Art. 2497-quater, para. 1(c)). However, the exit right is limited to non-listed companies. Furthermore, shareholders wishing to exit are required to show that they experienced an alteration in the riskiness of their investment ['alterazione delle condizioni di rischio'] as a consequence of the company's becoming part of the group. 44  offers stronger protection to the subsidiary 51 (i.e., to its minority shareholders and creditors), while the French and Italian approach offers more flexibility. 52 Before moving on, one additional feature of the regulation of groups in Germany, France and Italy must be highlighted. After the enactment of the Second Shareholders Rights Directive (SRD II), 53 IGTs involving EU-listed subsidiaries are subject to the special regulation of 'related party transactions' (RPTs) set forth in that directive. 54 Thus, IGTs involving a listed subsidiary are in principle governed by both group law and the special SRD II rules on RPTs. 55 This overlap raises some issues regarding how the ex ante controls on RPTs set forth in the SRD II should be performed when involving IGTs. We provide a brief account of these issues in Sect. 2.3 below, when we discuss the EMCA rules on groups.

The Call for Group Law from Advisory Expert Groups
The idea of providing corporate groups with special rules has been advanced by several expert groups, mostly comprised of legal academics, providing advice to, or with the intention of inspiring reforms by, EU policymakers. More precisely, three of these groups were set up by the European Commission (namely: the High Level Group of Company Law Experts, also known as the 'Winter Group' after the name of its chairman, Jaap Winter; the Reflection Group on the Future of EU Company Law; and the Informal Company Law Expert Group (ICLEG)) while three others were formed as spontaneous aggregations of legal academics.
The first among the latter was the Forum Europaeum Corporate Group Law, which proposed in 2000 the formal legal recognition of the interest of the group as a whole, 56 along the lines of the Rozenblum doctrine. 57 According to the Forum, managers of a subsidiary who operate 'in the interests of the group rather than in the commercial interests of the said subsidiary' 58 should not be considered as having breached their duties if the Rozenblum conditions are met. 59 Specific disclosures and detection mechanisms (most notably a special investigation procedure) 60 should apply in order to ensure compliance with the Rozenblum conditions. Support for a special group law (albeit expressed in somehow weaker terms than those of the Forum Europaeum) came in 2002 from the High Level Group of Company Law Experts. 61 The group advocated the view that there would be no need for 'an autonomous body of law, specifically dealing with groups' 62 at the EU level and that targeted interventions would be preferable. 63 Yet, according to the Group, ordinary company law rules may undermine group formation and functioning 64 and 'the acknowledgement of the legitimate nature of groups of companies necessarily implies that the company law rules on conflicts of interest and on the duty to pursue the sole interest of each company's shareholders cannot be applied as such to groups'. 65 Accordingly, the Group suggested the adoption of a 'framework rule' for groups that should allow 'those concerned with the management of a group company to adopt and implement a co-ordinated group policy, provided that the interests of creditors of each company are effectively protected and that there is a fair balance of burdens and advantages over time for each company's (outside) shareholder'. 66 The Group argued that such a regime 'would facilitate the creation and functioning of groups of companies'. 67 The Group did not provide indications about the content of this framework rule, arguing that 'there is a case for requiring Member States to provide for [such] a framework rule' 68 and that '[t]he details of the regime can be left to [the Member States]'. 69 It is quite clear, however, that this framework rule would have allowed Member States to recognise the interest of the group and permitted subsidiaries to adopt harmful decisions when they benefit the whole group.
A strong call for the introduction of special group rules came from the Reflection Group on the Future of EU Company Law. In its 2011 report 70 it urged EU policymakers to implement an EU rule offering explicit legal recognition of the interest of the group. The proposal was based on the idea that this recognition would, among other benefits, 'provide […] more clarity to the directors of the subsidiary as to which transaction or operations they can approve,' 71 thereby 'enhanc[ing] the flexibility of the management of groups especially on a cross-border basis'. 72 Another proposal in favour of special group regulation came in 2015 from the Forum Europaeum on Company Groups. 73 The proposed regulation would establish the subsidiary's duty to follow the parent's directions 'even if such directions are against a subsidiary's proper interests' so long as certain conditions-mostly 62 Ibid., at p 94. 63 Ibid., at pp 94-95. 64 Ibid., at p 96. 65 Ibid., at p 97. 66 Ibid., at p 97. 67 Ibid., at p 97. 68 Ibid., at p 97. 69 Ibid., at p 97. 70 See supra n. 35. 71 Ibid., at p 60. 72 Ibid., (italics in the original text). 73 Forum Europaeum on Company Groups (2015). This group has not only a similar name but also a similar composition to, and the same source of funding as, the one referred to supra n. 56 and accompanying text. See ibid., at p 299; Forum Europaeum Corporate Group Law (2000), at p 165. inspired by those of the Rozenblum doctrine 74 -apply. 75 Notably, however, minority-co-owned subsidiaries would have to have in place 'mechanisms concerning related party transactions of the kind provided in the recent proposals for reforming the Directive on Shareholders' Rights, if a transaction is substantial and not regularly occurring'. 76 A much more cautious endorsement in favour of the adoption of special group rules came in 2016 from the Informal Company Law Expert Group (ICLEG), 77 following the explicit request from the European Commission to address the matter. 78 After a detailed discussion of the merits of a special group law, 79 the group expressed the view that pan-European initiatives aimed at formally recognising the interest of the group should be limited to wholly-owned subsidiaries. 80 In the ICLEG's view, these initiatives could take the form of 'a provision allowing a wholly-owned EU subsidiary to recognise the interest of the group along the lines of a uniform formula'. 81 The ICLEG also prudently suggested that Member States should be left the option of extending a similar provision to minority-co-owned subsidiaries, 82 and that individual companies should also be left the possibility to opt into the special regime, provided that this decision is adopted via a supermajority shareholder vote. 83 Finally, unlike the other expert groups, the European Company Law Experts (ECLE), in their 2016 proposal for reforming group law in the European Union, 84 made no mention of the need to provide legal recognition of the interest of the group. Furthermore, the ECLE highlighted that group relationships are characterised by conflicts of interest 85 and should therefore be addressed through the rules governing RPTs. 86 Inspired by Member States' group laws (those of France and Germany above all) and by the ideas and suggestions of the expert groups mentioned above, EU policymakers have made multiple attempts to establish a pan-European legal framework for corporate groups.
German law on groups 88 and proposed to insert a detailed regulation on groups in its draft Regulation of the Societas Europaea. 89 Another attempt was made in 2003. In its Action Plan for the modernisation of company law, 90 the Commission, in line with the Winter Group proposal, 91 put forward the introduction of a 'framework rule' for groups, allowing for the adoption and implementation of coordinated group policies across affiliates. 92 To this end, it suggested the submission of a draft directive. 93 More recently, in its 2012 Action Plan, 94 the Commission endorsed the idea of 'an EU-wide move towards recognition of the concept of "group interest"', 95 anticipating a 2014 initiative in this respect. 96 To date, none of these attempts to introduce a special law of corporate groups has been successful. Accordingly, the EU currently lacks a harmonised special regulation for corporate groups along the lines of the national corporate laws of Germany, France or Italy.

The European Model Companies Act
The EMCA is a model law for a uniform European corporate law. 97 Inspired by a similar US model law, the Model Business Corporation Act, 98 and by the Principles of Corporate Governance drafted by the American Law Institute, 99 it was published in 2017 by a group of European academics. As a model law, the EMCA has no legal authority. It is mainly addressed to EU Member States as a template for future reforms of national corporate laws. Its ultimate goal is to promote further 'bottomup' convergence in an area of law that has traditionally showed resistance to full harmonisation. 100 While we are unaware of any national reform being explicitly drawn from the EMCA, this document is of interest for our analysis and not only because it 88 See Andersen et al. (2017). 98 Ibid., at p 1. 99 Ibid., at p 2. 100 Ibid., at p 1.
represents the latest scholarly proposal for a harmonised special regime for corporate groups. 101 In many respects, the EMCA consolidates previous efforts to articulate a pan-European regulatory framework for groups. For this reason, it may well inspire forthcoming group law reforms at both the EU and national level, including any initiative of the European Commission aimed at formally recognising the group interest in accordance with its 2012 Action Plan. 102 Under the EMCA, decisions contrary to the subsidiary's interest do not imply a breach of directors' fiduciary duties-and therefore may not give rise to directors' liability-provided that (i) the decision is in the interest of the group as a whole, (ii) the management, acting in good faith on the basis of the information that would be available to a diligent manager before taking the decision, may reasonably assume that the loss, damage or disadvantage will, within a reasonable period, be balanced by a benefit, gain or advantage, and (iii) the loss, damage or disadvantage suffered by the company in the first place does not put the company's continued existence in danger. 103 Under the same conditions, any parent's instruction to the subsidiary is to be considered as legally binding. 104 This special director duties regime is complemented by some measures in favour of minority shareholders: a sell-out right '[w]hen a parent company owns directly or indirectly more than 90% of the shares and of the voting rights of a subsidiary', 105 the right to appoint a special examiner in charge of assessing specific operations, 106 and the right to request a special investigation in the parent company in relation to decisions that affected the subsidiary. 107 To sum up, under the EMCA regime, directors of a group's subsidiary might adopt disadvantageous decisions that benefit the group (that is, the parent, another subsidiary, or each of the group's companies other than the harmed subsidiaries), provided that the harm so inflicted does not put the company's existence in jeopardy and, more importantly for our purposes, may reasonably be expected to be offset by a benefit, gain or advantage within a reasonable period.
The EMCA standard echoes the Rozenblum doctrine, 108 especially as regards compensation. Note that, differently from the national group laws discussed in Sect. 2.1, the EMCA does not require compensation to actually materialise ex post. It only requires that, at the time the harmful decision was taken, compensation could have reasonably been expected within a reasonable time. 109 The EMCA contains no indication as to how its special group regime coordinates with other parts of corporate law that may overlap or interfere with it, such as the general corporate law rules on shareholders' or directors' conflicts of interest. Most notably, the EMCA does not provide an explicit safe harbour that exempts group relations from rules on RPTs. This has significant consequences for EU-listed subsidiaries. As we have already noted, 110 after the enactment of the SRD II these companies are subject to harmonised rules on RPTs. According to them, inter alia, all material RPTs must be approved according to special procedures 'prevent[ing] the related party from taking advantage of its position and provid[ing] adequate protection for the interests of the company and of the shareholders who are not a related party, including minority shareholders'. 111 Nothing in the EMCA language suggests that IGTs, which plainly qualify as RPTs under the SRD II regime, 112 would be exempt from the application of SRD II rules on RPTs, and the SRD II, on its part, does not provide for any such exemption. 113 Thus, in principle, IGTs involving an EU-listed subsidiary will be subject to both EMCA group law (where adopted by a given EU jurisdiction) and the SRD II regime on RPTs.

European Group Law in Action
Many continental European legal scholars view special group law rules as highly valuable. 114 Such rules are thought to facilitate group management by providing group controllers with greater managerial flexibility, 115 because they relax the constraints imposed on group management by common corporate law director duties, 116 108 This is explicitly recognised by the EMCA drafters: see EMCA, at p 386. 109 This is similar to the Italian criminal law provision on offsetting advantages (see supra n. 41). As we will see in Sects. 3.2 and 3.3, this solution is not without consequences for minority shareholder protection against tunnelling. 110 See supra n. 54 and accompanying text. 111 See Art. 9c(4), SRD II. 112 The Directive adopts the same notion adopted by international accounting standards (see SRD II Art. 1, para. 2(b)(h)), according to which members of the same group are related parties. See International Financial Reporting Standards,IAS 24. 113 The SRD II does allow Member States to provide for an IGT exemption but limited to transactions between a listed parent and its own subsidiaries, provided that certain conditions are met. See Art. 9c, para. 6(a). Quite intuitively, these transactions pose no risks for the parent's minority shareholders. 114 See the positions of the many European expert groups recalled in Sect. 2.2. 115 See Conac (2013), at p 195; The Club des Juristes Committee on Europe (2015), at p 20; see also the position of the Reflection Group on the Future of EU Company Law, briefly summarised supra nn. 71-72 and accompanying text. 116 These rules, and especially the principle that directors must always act in the sole interest of their company, are thought to hinder some valuable and very common transactions in groups, such as cashpooling agreements and intra-group loans: see, e.g., Antunes et al. (2011), at p 61. especially their duty to act in the best interest of their company as variously labelled across jurisdictions (e.g., 'duty of loyalty' or 'duty to abstain or to deal fairly with the company in the presence of conflicts of interest'). The main operational implication and alleged shortcoming of the ordinary director duties regime in the group context is that they do not allow directors to properly take the broader group interest into account when making business decisions at the subsidiary level. 117 Decisions that are disadvantageous for the company but beneficial for the whole group imply a violation of these duties. In the view of group law supporters, this limitation unduly hinders efficient group management.
According to its supporters, group law would remove these frictions. In their view, group law would better accommodate the needs of firms organised as a web of legal entities under the same controller, increasing their aggregate value. 118 The alleged benefits of group law come mostly from its core principle, namely that the group interest may prevail over that of the affiliate, provided that the latter receives proper compensation for any harm suffered. In the EMCA variation, companies belonging to a group may take disadvantageous decisions (such as entering into unfair IGTs) that benefit the group as a whole, provided that the harm so inflicted can reasonably be expected to be compensated by a gain, benefit or advantage.
Not only would the EMCA standard provide the much sought-after managerial flexibility in group management, 119 but it would also allow for a more rounded assessment of the fairness of intragroup relationships, in the sense that it would reduce the risk of overreach (i.e., the tendency to unduly qualify IGTs as unfair), a risk that is inherent to ordinary rules against unfair self-dealing. 120 In the next sections, we examine these alleged benefits. We start by showing that the benefits usually associated with group law-even as epitomised by the flexible EMCA standard-are much more limited than group law supporters tend to think.

Increased Managerial Flexibility: Less Than Meets the Eye?
To understand the real contribution of the EMCA standard (and the similar national group laws across Europe) to group value maximisation, one must understand first how value maximisation may be achieved in groups and how ordinary director duties (and self-dealing rules more generally) may interfere with that goal. A closer look reveals that their interference is much less obtrusive than many (and especially group law supporters) tend to think. 117 Unless, of course, directors can bona fide conclude that it is in the company's best interest to do what is in the group's best interest, as established in UK law. See, e.g., Mevorach (2013), at pp 481-483. For a devastating critique of the very concept of 'group interest,' see Paz-Ares (2019), at pp 33-53. 118 See, e.g., Conac (2016), at p 311 (discussing the advantages of the EMCA principle and arguing that 'the possibility for the subsidiary to take into account the interest of the group allows for more flexible management of the group. The group can be more effectively managed as a single economic entity with a pooling of resources.'). 119 Ibid. 120 See generally Pacces (2019), at p 183 (arguing that substantive ex post court review of related party transactions tends to over-deter such transactions because of hindsight bias on the part of courts).
Group value maximisation may be achieved through Pareto efficient actions as well as through Kaldor-Hicks efficient ones. 121 Pareto efficient actions are those that increase group value without a negative effect on the involved affiliate's (or any other affiliate's) value. Examples include fairly-priced IGTs that allocate group resources (goods, services, and also business opportunities) to those group members who value them most and actions involving no formal exchange with other group members that increase group value by also producing net gains (or at least without bringing harm) at the single subsidiary level (e.g., investing in a positive net present value project that increases the affiliate's as well as the group's aggregate value). Kaldor-Hicks efficient actions are those that increase group value overall but negatively affect the single subsidiary. 122 One example of these is the use of unfairly-priced IGTs as a means to finance some positive net present value projects within a (cash-strapped) 123 group. That can be the case of what we term 'indirect intragroup financing,' where group affiliates are financed by other group members through favourable commercial terms in supply contracts rather than through a loan, such as when an input is purchased by a group member at below the price a competing independent supplier would charge. Another case in point is when a group affiliate undertakes a project that decreases the affiliate's value but has positive spill-over effects on other group members, outweighing the loss at the affiliate level.
How do ordinary director duties get in the way of such value-maximising actions? A plausible answer is that such duties cannot impede Pareto efficient actions (transaction costs aside: a point we fully address later in this section) but impede Kaldor-Hicks efficient actions (which are, by definition, unfairly priced). In fact, Pareto efficient actions (which are in turn, by definition, fairly priced) do not require directors to depart from their duty to always act in the company's best interest.Kaldor-Hicks efficient actions, on the contrary, imply that directors prioritise the group interest over that of the subsidiary or, in other words, sacrifice the affiliate's interest (e.g., by forcing it to enter into unfair transactions) in the name of the larger group interest. This intuition appears to underlie (and somehow represents a more accurate restatement of) the view, common among group law supporters, that ordinary director duties unduly hinder the efficient management of groups because they do not allow directors to properly consider the broader group interest when making a given decision. 124 However, a closer look at the concrete functioning of ordinary director duties reveals that they do not in fact prevent directors from adopting Kaldor-Hicks efficient actions. Ordinary fiduciary duties require directors to obtain compensation for the damage otherwise suffered as a consequence of the relevant decision, as a 121 See Denozza (2000), at p 329. 122 Of course, for the transaction to be efficient (i.e., value-creating) the benefits for the whole group must be larger than the loss experienced by the subsidiary. As clarified below (see infra text following n. 124), transactions that would otherwise be merely Kaldor-Hicks efficient can be designed as Paretoefficient ones by requiring another group member to fully compensate the subsidiary for the harm deriving from the transaction's own terms. 123 See also infra n. 140. 124   condition for legitimately taking that decision. Since Kaldor-Hicks efficient actions, by definition, create net additional wealth, there should always be room for such a negotiating outcome. Notably, compensation might also be deferred (i.e., a legally enforceable promise of future compensation may also suffice to exclude director liability for breach of fiduciary duties). 125 It follows that the EMCA standard does not enable directors to pursue groupvalue-maximising strategies which traditional ordinary director duties would rule out. The main difference is that under traditional director duties regimes, the directors who fail to obtain compensation from the parent are liable for breach of their duty of loyalty, whereas under the EMCA standard they can argue that at the time the transaction was entered they could reasonably assume that the harm would be offset by a benefit within a reasonable time. In other words, not even under traditional fiduciary duties the terms of the transaction must be per se fair. Yet, it remains a duty upon the directors to make sure that compensation duly ensues, the violation of which gives rise to their liability unlike under the EMCA standard.
Yet, there might still be advantages to the EMCA approach. As we show below, the EMCA standard allows actions increasing value at the group level to be undertaken at a lower cost relative to what would be required according to ordinary corporate law, thereby permitting more of this value-creating activity. However, as we argue below, the cost reduction is less relevant than one might think and the ensuing marginal benefits are correspondingly lower. Let us start by explaining how the EMCA standard might reduce the (transaction) costs of undertaking actions that increase group value.
Consider first group value maximisation via the efficient allocation of resources across the group. Ordinary corporate law may impose (and frequently does impose) some costs for entering into these transactions. Specific approval procedures often apply that aim at ensuring that these transactions, tainted by conflicts of interest, are fair to the company (for instance, that the consideration paid is not higher than what the asset acquired is worth to the company). Depending on the specific contents and features of a jurisdiction's corporate law (both on the books and in action), these controls can be costly and, therefore, at the margin, may prevent some IGTs from being entered into. 126 That will more likely be the case where fair-dealing rules are 125 Furthermore, under ordinary fiduciary duties compensation is likely not required where the purpose of the harmful action is the avoidance of a greater disadvantage (e.g., the harm ensuing from the collapse of the entire group). This principle, explicitly established in English case law (see Charterbridge Corp v. Lloyd's Bank [1970] Ch. Div. 62, at 74-75), is a corollary of the very duty to pursue the company's best interest. Indeed, undertaking a harmful action (e.g., entering into an unfairly priced IGT, such as when a guarantee is provided for free) that avoids an even greater harm can be held to be an action in the best interest of the company. See Davies et al. (2021), at pp 280-281. 126 As an example, consider the case of company A holding an asset that it values at 9.5 million euro, because it would be able neither to sell it for more to any non-related entity nor to use it in its production process to create a profit higher than that amount in present value terms. Suppose that sister company B values the asset at 9.6 million euro. If the costs for A of complying with the self-dealing regime when entering the IGT transferring the asset are higher than 100,000 euro, then the transaction will not be entered into and B will be unable to exploit the asset as the highest value user. stringent and where access to justice for aggrieved minority shareholders is easy and cheap.
Assume that corporate law establishes that conflicted transactions must be reviewed and approved ex ante by an independent decision-maker (as required of all EU-listed subsidiaries following the enactment of the SRD II 127 ), such as a committee of independent directors. This obviously entails costs. The time required to approve the transaction in many cases will be longer than that required to make a non-conflicted business decision, and the decision-making process might involve higher expenditures (e.g., by de facto requiring costly third-party advice). In addition, if independent directors have full authority over the IGT, but (as is reasonable to assume) less information than the executives, they may hold out in the wrong but sincere belief that the terms proposed do not reflect what the asset bought or sold is worth to the company, leading the company to also forgo fair and value-increasing transactions. In other words, the independent directors' inferior firm-specific knowledge may lead to valuable transactions not being entered into. 128 Costs are also imposed on the transacting parties if the transaction is subject to the approval of unconflicted shareholders. Shareholder involvement in the decisionmaking process carries both direct costs (e.g., a shareholder meeting must be called, and detailed information must be prepared and distributed in advance to shareholders) and indirect ones (e.g., an informed shareholder vote may require disclosure of sensitive business information, thereby reducing the company's profitability). 129 In addition, despite the information costs often incurred by the company and its shareholders, there still exists the risk of a wrong assessment of the transaction (i.e., of shareholders mistakenly blocking advantageous transactions), given outside shareholders' inferior company-specific knowledge.
Finally, policing self-dealing not via ex ante controls but rather in the form of ex post judicial review of a transaction's procedural fairness also entails costs. With time, if the Delaware experience is instructive in this regard, procedures similar to those illustrated above (e.g., approval by independent directors and/or approval by a majority of the minority shareholders) may emerge as hallmarks of procedural fairness, leading to a similar outcome. 130 In the process, if litigation is easily and cheaply accessed, litigation costs will be incurred, the expectation of which may, at the margin, have a chilling effect, ex ante, on (fair) value-creating IGTs (but, needless to say, on unfair ones too).
As an outcome, in groups where IGTs are routine (as in the case of verticallyintegrated groups with affiliates operating at different levels of the production 127 See supra n. 111 and accompanying text. 128 This risk appears particularly high with respect to transactions having idiosyncratic features (like those involving the sale or purchase of unique goods, such as customised components for the production of complex goods), for which there are no reliable market benchmarks against which to assess the fairness of the price paid or received. See Pacces (2019), at pp 183, 196-199 (stressing the limits of marketbased criteria in screening RPTs); see also Gözlügöl (2022), at pp 77-86 (proposing a number of alternative criteria to screen the fairness of IGTs with idiosyncratic features). 129 Enriques (2015), at p 18. 130 See, e.g., Enriques et al. (2017), at pp 154-155. chain), the costs imposed by self-dealing rules may be non-negligible, thereby possibly posing an obstacle to group value maximisation. 131 To be sure, group controllers may avoid these obstacles by restructuring their group so as to merge all subsidiaries into one company or by buying (or, where available, freezing) out minority shareholders. This choice, however, would itself be costly: first, it entails giving up the benefits that the MCOG ownership structure may yield. 132 Second, because the reorganisation itself would be executed via one or more RPTs (e.g., a parent-subsidiary merger), it would entail the costs of entering these. It would therefore be a viable option only where the benefits (namely the prospective reduction in transaction costs) outweigh the costs of the transition to a single-entity firm.
The EMCA standard, while being silent on the approval process for IGTs (which would then be the one applicable according to general self-dealing or RPTs rules), would still reduce the costs of intragroup exchange. It would do so by curbing IGTrelated litigation. In fact, the defence that the EMCA establishes in favour of subsidiary directors would make it harder for minority shareholders to challenge even unfair IGTs. 133 As a consequence, fewer IGTs would be challenged in court, thereby reducing expected litigation costs, which in turn can be considered as part of IGTs' transaction costs. To the extent that the risk of litigation is lowered, group subsidiaries can be expected to act in ways that decrease the effectiveness, and therefore the impact and costs, of the procedural safeguards to be applied to IGTs qua selfdealing transactions. 134 For instance, in anticipation of the even rarer use of minority 131 Corporate law rules against unfair self-dealing are clearly designed for stand-alone companies, which by their very nature only episodically engage in transactions with related parties. For the very reason that such firms are not part of an integrated firm, stand-alone companies' RPTs justify greater suspicion. In an important sense, these rules are designed to discourage companies from engaging in related party transactions altogether, limiting self-dealing to those cases in which no viable alternative exists. Cf. Davies and Hopt (2013), at pp 352-353 (highlighting how the result of regulations on RPTs creating high procedural hurdles to these transactions 'may ultimately be the same as a substantive prohibition'). Notice, however, that policymakers have devised ways to reduce the impact of self-dealing rules in companies where IGTs are frequent. The Italian regulation of related party transactions allows companies to reduce the number of times RPTs must be subjected to independent directors' review by allowing 'framework resolutions' ('delibere-quadro') for the approval of homogeneous RPTs. Under this arrangement, homogeneous RPTs can be grouped for the purpose of independent director review, thus avoiding the costs of applying self-dealing controls to each and any of such transactions. See Consob, Regolamento n. 17221/2010 (Operazioni con parti correlate), Art. 12 (for a convenience translation see https:// www. consob. it/ web/ consob-and-its-activ ities/ laws-and-regul ations/ docum enti/ engli sh/ laws/ reg17 221e. htm? hkeyw ords= & docid= 11& page= 0& hits= 24& nav= false). This rule makes it easier for group firms to implement arrangements or strategies that entail frequent intragroup exchange, such as cash-pooling agreements. 132 See Enriques and Gilotta (2024, forthcoming) 133 We show in Sect. 3.3 how this feature of the EMCA standard may impair minority shareholder protection against tunnelling. 134 Consider that the introduction of a principle exempting directors from liability when they adopt a disadvantageous decision that benefits the group in the expectation of a future offsetting advantage does not automatically exempt directors from compliance with procedural rules on IGTs qua self-interested transactions. To this end, an explicit exemption must be established, which the EMCA (as we already noticed: see supra text accompanying nn. 112-113) does not contain. However, as the text clarifies, it is also worth noting that the introduction of that principle may affect the way ex ante controls on RPTs (where required) are performed, de facto leading to screening criteria more favorable to controllers. Indeed, the existence of a principle that exempts directors from liability for unfair transactions for which compensation could reasonably be expected may support the claim that, as a matter of law, independent directors shareholder remedies, less information is likely to be provided to independent directors in charge of vetting IGTs, so as to minimise the risk that they raise objections. More generally, compliance with procedural fairness rules and practices may be weakened, with the effect of reducing the overall contracting costs of IGTs.
The EMCA's granting of an explicit permission to make decisions that are harmful to the individual group entity also makes it easier for controllers to enter into Kaldor-Hicks-efficient indirect intragroup financing arrangements. 135 Under ordinary corporate law, not to be held liable for breaching their duties, directors of the financing company must obtain a formal (i.e., legally enforceable) promise of indemnification from the financed company or from other group members (typically the parent company). The indemnification agreement, in turn, would seem to require that the harm inflicted on the subsidiary be quantified ex ante. By contrast, the EMCA standard appears to require neither a formal indemnification promise nor quantification of damages. The indemnification promise appears not to be necessary because directors are explicitly allowed to deviate from their duty to act in the company's best interest and enter into an unfair transaction in the first place so long as they can reasonably expect future compensation. The condition that future compensation may reasonably be expected at the time the unfair transaction is entered into, in turn, likely requires that directors be aware of the harmful nature of the transaction and, perhaps, that they have a rough idea of the harm's magnitude. However, the EMCA standard falls short of requiring them to either explicitly and precisely quantify harm or identify the offsetting benefit, gain or advantage that will neutralise it ex post. Thus, the standard reduces the risk that efficient indirect intragroup financing arrangements are forgone due to quantification errors ex ante, or judicial errors ex post, thereby implying lower transaction costs for indirect intragroup financing under the EMCA standard than under ordinary self-dealing regimes.
Finally, we already pointed out that ordinary fiduciary duties do not prevent directors from undertaking actions (other than IGTs) that harm the subsidiary but yield a higher gain at the group level thanks to spill-over effects and, thus, maximise group value. Similar to the case of indirect intragroup financing, those duties rather require them to obtain compensation or a legally enforceable promise thereof for the harm 135 See supra text following n. 123.
Footnote 134 (continued) charged with the task of vetting IGTs are allowed to-and perhaps even should-approve unfair transactions for which compensation may reasonably be expected. Some national corporate laws touch upon these issues. The Italian regulation of RPTs in listed companies, for instance, clarifies that procedural and transparency rules on related party transactions also apply to IGTs. It also attempts to coordinate such rules with group law principles when both come into play. It does so by stating that, in companies subject to the direction and coordination of another company (a de facto situation which is presumed to exist in the presence of control by another entity and therefore in the case of MCOGs), the independent directors' advice on RPTs shall indicate why the relevant transaction is fair to the subsidiary, 'as the case may be, also in light of the overall result of the activity of group direction and coordination or in consideration of transactions aimed to fully offset the damage resulting from' the relevant RPT, therewith using almost the same wording (in italics here) as in Art. 2497, Civil Code (see supra n. 41). See Consob, Regolamento n. 17221/2010, supra n. 131, Art. 14. See also Gilotta (2012), for a discussion of the interplay between rules on RPTs and group law in Italy.
inflicted on the company. Requiring indemnification generates contracting costs that inevitably impede some of these bargains (those where the gains do not exceed the higher contracting costs). The requirement for an indemnification agreement turns what was a decision entailing no formal transaction with other group members into a formal IGT whereby the company 'trades' its choice to invest in (forgo) a negative (positive) net present value project with the promise of a sum of money or other benefit offsetting the loss suffered. If a formal IGT must be entered into, the costly rules against unfair self-dealing (e.g., independent directors' approval) become applicable, impeding all transactions whose gains do not outweigh the costs generated by those rules. Furthermore, similar to what we have seen above with respect to indirect intragroup financing, the need for an explicit indemnification agreement would seem to require directors to explicitly quantify the harm ex ante, which in turn increases the risk of false positives. The EMCA standard, on the contrary, allows directors to implement these actions with no need to turn them into costlier IGTs.
The EMCA standard thus reduces the costs associated with the implementation of group-value-maximising actions to the extent that it reduces the likelihood of litigation on intragroup exchange. As anticipated, though, this cost reduction may easily be overstated.
First of all, the EMCA standard is unlikely to provide any significant cost reduction wherever litigation on intragroup exchange is rare anyway. This is the case for most European jurisdictions, where a number of obstacles (such as limited inspection rights, restrictions on standing to sue and derivative suits, and limits to contingency fees) prevent minority shareholders from systematically (if not even occasionally) challenging conflicted transactions before courts. 136 Put differently, the EMCA standard reduces the risk of something that is already highly unlikely, which implies that the ensuing reduction in contracting costs cannot be significant.
Secondly, a similar, but more specific argument undermines the idea that the EMCA standard may reduce the costs of indirect intragroup financing. We have already pointed out that the EMCA decreases the contracting costs of this financing technique, relative to general corporate law. However, for this advantage to be meaningful, groups would have to heavily resort to indirect intragroup financing to fund their positive net present value projects. At least for those operating in welldeveloped financial markets, like those in many of EU jurisdictions, this appears to be implausible. 137 Indeed, groups operating in (or otherwise having access to) 136 For France, Germany and Italy see, e.g., Conac et al. (2007), at pp 508-509; Gelter (2011), at pp 856-880. See also Gerner-Beuerle and Schuster (2014), at pp 214-222, 233 (providing a detailed account of the obstacles minority shareholders face across EU jurisdictions in enforcing director duties and arguing that no European jurisdiction provides minority shareholders with a regulatory environment strongly and fully conducive to minority shareholder suits). And see infra n. 154. Note, however, that U.S. courts appear increasingly open to hearing derivative suits brought against foreign companies (see Blake et al. 2021), including European ones (see Smith (2021)). If this trend were to continue and to strengthen, it might eventually make minority-shareholder-driven litigation (over conflicted transactions and other issues) less infrequent. 137 We were unable to find any empirical evidence about the relevance of this phenomenon that may belie the statement in the text. Indirect intra-group financing may instead provide significant benefits where financial markets are underdeveloped, as it may help firms overcome the hurdles that they encounter in obtaining external finance. See, e.g., Khanna and Palepu (2000), at p 868, for the general (and developed financial markets in principle do not need to resort to indirect intragroup financing to fund their positive net present value projects, since to this end they may easily resort to external finance. Group law supporters may object that in a competitive and efficient capital market, resorting to the group's internal capital market is in principle cheaper than resorting to external finance, since the latter entails transaction costs, problems of contractual opportunism and informational asymmetries that are much less consequential when the lender and borrower are, economically speaking, part of the same firm. 138 Yet this objection overlooks the fact that indirect intragroup financing has its own indirect costs: it is an opaque and rather convoluted financing technique that, in requiring systematic deviation from fair exchange in IGTs, is likely to raise concerns among minority shareholders, outside creditors and the larger financial community. 139 Under the reasonable assumption that in a developed and competitive financial market external finance is relatively cheap (though, admittedly, not as cheap as internal capital market financing), those costs are likely to make indirect 138 A large strand of the finance literature stresses, from either a theoretical or empirical perspective, the advantages of internal capital markets vis-à-vis external capital markets: see, e.g., Shin and Stulz (1998), at p 531, for the general observation that '[t]he internal capital markets of diversified firms enable them to fund profitable projects that, because of information asymmetries and agency costs, the external capital market would not be able to finance'; Gertner et al. (1994) (arguing that internal capital market financing allows for better debtor monitoring and easier asset redeployment than external bank financing); Hoshi et al. (1991) (finding that group affiliation eases firm financing constraints by making investment less sensitive to liquidity). In addition, indirect intragroup financing implies funding the 'borrowing' company from the 'lending' company's surplus (because until the offsetting advantage materialises, the lender's profits are lower than they would otherwise be), which is cheaper than resorting to external debt (see generally Myers and Majluf (1984)). Furthermore, the advantages of indirect intragroup financing over external financing may be expected to be larger with respect to small-or medium-size non-listed subsidiaries, because of the larger informational asymmetry between them and external investors. Note, however, that internal capital markets have their own distortions. See, e.g., Scharfstein and Stein (2000) (showing that weaker subsidiaries tend to be subsidised by stronger ones); Rajan et al. (2000) (showing that when group divisions diverge in resources and opportunities, internal funds tend to flow to the most inefficient divisions); Almeida and Wolfenzon (2006) (arguing that resource allocation via groups' internal capital markets, even when efficient from the group's perspective (in terms of picking the most profitable projects), leads to economy-wide inefficiencies, preventing the efficient allocation of resources in the economy). See also Kabbach de Castro et al. (2022) (showing that group affiliation and the subsequent recourse to the group's internal capital market may not always mitigate the financial constraints that firms encounter when resorting to external finance). 139 It may be worth recalling that indirect intragroup financing is the financing of a group affiliate via unduly advantageous IGTs (e.g., the sale of input for less than its market price) that harm the counterparty (in our framework, the affiliate with minority shareholders). As a financing technique, indirect intragroup financing is thus different from-and much more opaque than-'direct' intragroup financing, namely the financing of group affiliates via intragroup loans, share purchases, or other 'explicit' financing contracts.
Footnote 137 (continued) widely shared) observation that groups may play a valuable gap-filling role where external markets are absent or underdeveloped. In addition, the marginal value of intra-group financing in general may be higher in exceptional times, such as a pandemic or (the aftermath of) a financial crisis, when, on the one hand, lenders may refuse credit for liquidity or regulatory reasons and, on the other, firms throughout the economy may have liquidity or solvency problems. That is why one of us has suggested to suspend the (mandatory) application of procedural rules on RPTs in exceptional times such as a pandemic. See Enriques (2020), at p 266.
intragroup financing a truly valuable option only in exceptional circumstances. 140 Accordingly, the benefits of group law cannot be expected to be significant for such transactions either. To sum up, the reduction in contracting costs that the EMCA standard would entail would apply to transactions that groups can be expected to undertake infrequently.
Finally, consider value creation via non-transactional actions or omissions that harm the subsidiary but are value-creating at the group level thanks to intragroup spill-overs. One obvious strategy here is that of requiring subsidiaries to forgo profitable expansion into related business areas in which other group members successfully operate. This strategy maximises the group's profits by minimising intragroup competition. 141 We showed that group law allows these actions to be executed at a lower cost relative to ordinary corporate law. At the same time, it must be noted that the chances of directors being held liable under ordinary corporate law for this kind of actions are negligible. Consider the case of a subsidiary's directors that forgo a profitable business expansion so as not to cut into the profits of other group members. If directors, as can be reasonably assumed, do not adopt any formal resolution not to undertake the project but simply shy away from even considering the matter, the chances that, even under ordinary corporate law, they will be held liable for violation of their fiduciary duties in a subsequent shareholder-initiated lawsuit are virtually non-existent, since there is no formal decision or act to which their liability can be attached. 142 Thus, in this case the benefits of group law are also more theoretical than real.
To conclude, the EMCA standard allows group-value-maximising actions to be undertaken at a lower cost relative to ordinary corporate law, and there resides its efficiency contribution. Yet, the size of this cost reduction is in practice not particularly significant, at least so long as the group has access to developed capital markets and shareholder-driven litigation over director fiduciary duties is rare. Thus, at least in Europe, the net efficiency benefits of the EMCA standard can be expected to be limited.

There May Be Fewer False Positives, But What About False Negatives?
In the view of the EMCA standard's supporters, general corporate law rules against unfair self-dealing are also ill-suited for groups because they fail to provide an accurate assessment of fairness in intragroup exchange. 143 These rules are usually 'transaction-centred': they focus on whether a single conflicted transaction, like an IGT, is fair (e.g., whether the company paid or received a fair price for what it bought or sold in that individual transaction). 144 This focus on the single transaction-the argument goes-fits poorly with the reality of intragroup exchange, 145 often characterised by a sequential give-and-take logic of disadvantageous (i.e., unfairly-priced) transactions followed (or, as the case may be, preceded) by (usually implicitly) compensatory transactions.
According to this view, standard self-dealing regimes would inevitably be overinclusive: they would deliver a disproportionate number of 'false positives', where IGTs are mistakenly considered unfair because of the rules' failure to assess a transaction's fairness in light of other benefits or advantages that the subsidiary will subsequently receive. 146 The EMCA standard, like some European national group laws, would avoid this problem with ordinary rules against unfair self-dealing because it would allow a transaction to be evaluated in light of unrelated (and possibly even just reasonably expected) benefits or advantages. The inclusion of offsetting benefits would, in this view, deliver an overall more reliable assessment of fairness in intragroup exchange, leading to an overall lower number of false positives. The underlying intuition is that unfair decisions (e.g., unfairly-priced IGTs), for which compensation of harm can reasonably be expected to occur, are ultimately innocuous decisions that do not cause value diversion. The EMCA provides that these transactions do not give rise to directors' liability and hence removes a hindrance to entering into them. By reducing the number of value-increasing transactions thus impeded by self-dealing rules, a more accurate assessment of intragroup exchange would promote value creation. We have already seen how concerns over directors' liability may be overblown in countries where derivative suits are exceedingly rare. 143 See Dammann (2019), at pp 218, 232 (stressing how 'requiring individual transactions to satisfy arm's length standards without considering overarching benefits and costs created by the corporate group may prove problematic'). 144 Fair price, in turn, is usually defined as the price that the company would have agreed to receive or pay after a hypothetical arm's length negotiation with an unrelated third party. 145 Dammann (2019), at p 233 (arguing that 'a system that focuses on individual transactions may be both too strict in some cases and too lenient in others'). 146 See Dammann (2019), at pp 232-233 (stressing in a similar vein that standard self-dealing rules may discourage the formation of efficient groups because of the rules' failure to factor in the benefits that the group structure provides to the single affiliate).
To be sure, there might be other negative legal consequences that directors may want to rule out when deciding on IGTs, 147 which may make the EMCA standard, if applicable across the board, useful when addressing the false positives problem.
Yet, the beneficial effect of reducing false positives goes together with a second predictable effect of a loose standard such as the EMCA, namely that a higher number of false negatives (i.e., value-diverting transactions mistakenly judged not to be value-diverting) should similarly be expected. 148 Errors in the assessment aimed at verifying whether, faced with an unfair transaction, it had been reasonable for directors to expect that the harm would be compensated may indeed lead to a higher number of tunnelling transactions being mistakenly judged to be fair than would be the case in the absence of the EMCA standard.
But, worse still, even the correct application of the EMCA standard may lead to false negatives. In fact, defendants may well persuade the court that, when the IGT was decided upon, it was reasonable to expect intragroup compensation regardless of the fact that no intragroup compensation did ex post materialise. Of course, plaintiffs could argue that the fact that no intragroup compensation has been shown to have taken place is indirect evidence of the unreasonableness of the expectation. But it would be far from a dispositive argument, especially if directors were wise enough to make sure that convincing explanations for the reasonable expectation of future compensation were duly recorded, for instance in the minutes of the board meeting in which the transaction was approved.
Clearly, if false negatives (transactions harming the subsidiary without compensation) become more likely, controlling shareholders will have stronger incentives to enter into unfair IGTs. A situation could even arise where they may well enter a transaction whenever they stand to gain from it, even though the transaction itself destroys value and actually reduces group wealth.
Group wealth reduction occurs when, for example, an asset is transferred from a minority-co-owned subsidiary to one wholly owned by the parent and where the latter values the asset less than the former. The group's controllers may well be happy with the transfer if, in doing so, they extract private benefits that are higher than the pro quota reduction in the value of their equity investment in the selling subsidiary.
To illustrate this point, suppose that asset X is transferred from (minority-coowned) subsidiary A (controlled by B with 50 percent of the shares) to subsidiary C (wholly owned by B) at a price of 70, where the asset is worth 130 to A and 110 to C. B's loss from the transfer qua shareholder of A is 30 (half of 130 minus 70) but her gain qua shareholder of B is 40. In the process, the aggregate value of the group has gone down by 20 because C values X 110. Importantly, this simple example also shows that it is wrong to assume that the interests of the decision-makers at the top of the group will be aligned with those of other shareholders of the group's subsidiaries. 149 In the presence of such a wedge, it is far from certain that the net outcome of a laxer standard such as the EMCA will lead to an increase in the value of business groups as a whole and, therefore, to efficiency gains. Given that the final decision-making power within groups rests with the controlling shareholder, it is not intuitive, to say the least, that the overall outcome of intragroup exchange is value creation. 150 To be fair, EMCA's requirement that the relevant decision be 'in the interest of the group' means that, though harmful for the affiliate, its outcome must be valueincreasing at the group level (i.e., it has to be at least Kaldor-Hicks efficient 151 ). Yet there can be no certainty that a court will be able to always distinguish behaviour that creates value at the group level from behaviour that destroys it. Under the EMCA standard, when a court errs in that assessment and wrongly judges the relevant behaviour to be in line with the interest of the group, directors will face no liability if they persuade the court that it was reasonable to expect offsetting advantages at the time the decision was taken. If ordinary corporate law applies, it is irrelevant whether the behaviour creates or destroys value at the group level: if it harms the subsidiary, its directors will be held liable. Hence, under the EMCA standard the controlling shareholder may more easily get away with value-destroying behaviour than under ordinary corporate law rules.

Why the EMCA Standard Would Be a Free-for-All
Under the EMCA standard, there would be an easy and effective way for defendant directors to corroborate, if not prove, the claim that it was reasonable to expect, at the time the decision was made, that the harm would be offset at a later stage by a benefit, gain or advantage. They could show that, as they had expected at the time of the relevant decision, the company did later receive benefits from the group and that those benefits were valuable enough to offset the harm caused by their decision. The reason why this defence might easily work is two-fold. On the one hand, it would be unpersuasive to allege that an expectation of future compensation was unreasonable ex ante when such compensation did materialise ex post. On the other hand, consider that existing national group laws often require actual compensation in order to exempt directors and/or the parent from liability. 152 Although the EMCA standard is designed as an ex ante, rather than as an ex post, standard, in applying it national courts might de facto lean towards the same approach followed under the national group laws they will have previously applied for decades.
If this defence were to work, it would practically sanction tunnelling other than in the most egregious cases. In fact, were a harmful action (e.g., an unfair IGT) challenged in court, it would not be difficult for controllers to identify ex post a quantifiable advantage of some sort that the subsidiary later received from another group entity (if not from being part of the group itself 153 ) and present it in court as the ex post compensation for the specific harmful action minority shareholders are suing the directors for. Yet, the decisions that are actually challenged in court will usually represent a fraction of the harmful actions a subsidiary may undertake: many, if not most, of them will simply go undetected, given the limited access to corporate information minority shareholders have in European jurisdictions. 154 For others, unfairness may be too difficult to prove. In addition, hurdles to private enforcement may prevent minority shareholders from initiating a lawsuit. 155 As a consequence, a 'fractional reserve' of group-related benefits will likely suffice to prove that directors' expectations about offsetting benefits had been reasonable at the time of the 153 N that in applying group law, Italian courts do not admit defenses based on the generic assertion that adequate offsetting stems from the very fact of being part of the group. Instead, according to Italy's Supreme Court, defendant directors have to prove that specific benefits exist that are fully capable of providing adequate compensation. Yet this does not imply that defendant directors may not resort to benefits that the company obtains from its participation to the group as a whole, rather than from one or more group members. See, e.g., Cassazione civile, sez. I, 24/08/2004, n. 16707, in 32-II Giurisprudenza commerciale 246 (2005). 154 Access to information is bound to be limited whatever the general and RPT-specific mandated disclosures are and despite the inspection rights available to minority shareholders whether according to EMCA (see supra nn. 106-107 and accompanying text) or national company laws (see, e.g., Conac et al. (2007), at pp 512-513). In particular, mandatory disclosures about IGTs are unlikely to always contain enough information to enable minority shareholders to assess a transaction's fairness. The SRD II requires Member States to 'ensure that companies publicly announce material transactions with related parties at the latest at the time of the conclusion of the transaction', and establishes that '[t]he announcement shall contain at least information on the nature of the related party relationship, the name of the related party, the date and the value of the transaction and other information necessary to assess whether or not the transaction is fair and reasonable from the perspective of the company and of the shareholders who are not a related party, including minority shareholders' (Art. 9c, para. 2). Non-material transactions are thus exempt, and non-transactional intragroup exchange is obviously not covered. In addition, in some Member States routine transactions are subject to reduced disclosure obligations: see, e.g., the Italian regulation on related party transactions: Art. 13, para. 3(c), Consob Regulation No. 17221/2010. EMCA provides minority shareholders with inspection rights to be exercised at both the affiliate and the parent level (see supra nn. 106-107 and accompanying text). Yet the 10% share ownership threshold required to exercise those rights appears prohibitively high, at least for listed subsidiaries, where minority shareholders will find it hard, if not impossible, to reach the threshold. Furthermore, the inspection is carried out by a court-appointed examiner whose incentives might not always be aligned with (or as strong as) those of minority shareholders and their lawyers. Finally, consider that discovery during trial or at the pre-trial stage is traditionally unavailable in continental Europe. See, e.g., Giudici (2004) decision. In other words, controlling shareholders will know that any benefit x a subsidiary can be held to have received from intragroup relationships unleashes the possibility of harm n times x to the same subsidiary, where n > 1.
Anticipating this, courts may decide to admit ex post compensation as indirect evidence of compliance with the EMCA standard only if offsetting benefits were proved to have been received within a given ('reasonable') time span. This solution would be similar to the German standard on intragroup exchange, 156 except for the absence of a precise timeframe. 157 One may object that the EMCA's language refers to a benefit, gain or advantage, suggesting that for the defence to work it should be sufficient for directors to show that the company received some benefit, and that that benefit outweighed the harm; hence, as the argument would go, the EMCA requires something less than the German-like solution sketched out above. Yet, assuming for a moment that minority shareholders' access to corporate information was unlimited, the practical implications of this reading of the EMCA would not be significantly different than under a German-like standard. Once directors show the existence of an offsetting benefit, plaintiff minority shareholders should be allowed to allege and prove other harmful actions neutralising the benefit identified by the defendants, to which defendants could in turn respond by mentioning additional offsetting benefits. A point could thus be reached where the facts of the case extend to the entire set of intragroup interactions within an undefined (but reasonable) timeframe (i.e., the German-like solution again).
Yet, once we relax the assumption that plaintiff minority shareholders have full access to corporate information, even this German-like solution reveals itself to nicely serve the defendants' interests: it will be easier for them to find offsetting benefits than it will be for plaintiffs to find and prove additional harmful actions.
Tunnelling risks would increase further if courts were to accept the relevance of the group's record of intragroup exchange up until the time the decision was taken, in order to determine whether directors' expectation of future compensation was reasonable. A similar approach is in fact followed by French courts in applying the Rozenblum doctrine. 158 Assume that for the past 5 years company A sold raw materials to sister company B for less than their market value, each year receiving 156 German Konzernrecht requires that all IGTs executed in one financial year be checked by an independent third party in order to verify whether on balance value diversion occurred as a consequence of those transactions. See Aktg, § 313, para. 1(2). See also supra n. 30. Recent rulings from Italian courts applying Italy's rules on groups seemingly hint at the solution in the text. The criminal section of the Italian Supreme Court, in a number of recent cases, established that in order to prove the existence of offsetting advantages the defendant must show a positive net balance ('saldo finale positivo') from the transactions carried out in the interest of the group. See, e.g., Cassazione penale, sez. V, 30/06/2016, n. 46689, Rivista dei dottori commercialisti 321 (2017-2), at p 321; Cassazione penale, sez. V, 27/02/2020, n. 13284, Guida al diritto 83 (2020-34/35), at 84. Note, however, that these court rulings are far from being settled case law. 157 See supra n. 44 and accompanying text. 158 As reported by Pierre-Henri Conac, 'French courts assume that there will be a quid pro quo for a disadvantageous transaction based on the previous or supposed behaviour of the group towards its subsidiaries'. See Conac (2020), at p 92 (emphasis added). marketing services 6 months later from sister company C at a discounted price that fully compensated the loss incurred from the supply contract with B. On the basis of this track record, A's directors' expectation that selling raw materials today to B for less than their market value will be compensated ex post may well be found to be reasonable.
All of this would create a strong incentive for end-game tunnelling: once a track record of past compensation between A and B has been established, but B no longer needs to purchase the input from A (e.g., because that input will start being provided by another group member or by an external supplier, a decision which is of course under the controlling shareholder's control), controllers may be induced to change the terms of the compensating transaction to the detriment of A.
Note, finally, that the EMCA is silent on the issue of whether, in the absence of ex post compensation, minority shareholders may otherwise obtain indemnification from the parent or from the group members which actually benefited from the harmful action. 159 Unless other more specific corporate or private law doctrines apply that qualify the parent's behaviour as giving rise to liability or indemnification obligations, minority shareholders may have to resort to residual remedies, such as unjust enrichment, the availability of which will depend on their specific contours according to the applicable national private law regime. Obtaining restoration through these remedies may be even more difficult than through ordinary corporate law remedies such as derivative suits against directors.
To sum up, challenges to controllers' value diversion would become significantly more difficult for minority shareholders under the EMCA. Proof that a given decision was harmful to their company would not suffice to affirm directors' liability. In fact, directors could successfully rebut this allegation by proving that it had been reasonable for them to expect that compensation would follow, possibly by producing evidence that an offsetting benefit, gain or advantage did actually materialise. 160 When such evidence is produced, minority shareholders would have a hard time proving that those offsetting advantages were in fact compensation for other harmful actions.
Clearly enough, such a legal framework would be much more conducive to tunnelling, thus significantly decreasing the protection offered to minority shareholders in groups. While the direct effects of tunnelling are distributional, as pointed out before, tunnelling also has efficiency repercussions, in terms of higher agency costs 159 EMCA does not address this issue (see EMCA,sect. 15,. This marks a difference with national groups laws. German group law establishes the parent company liability toward minority shareholders of the subsidiary for damages suffered as a consequence of uncompensated harmful actions. See AktG, § 317. Italian group law also explicitly affirms the parent's liability for such damages. However, a lawsuit against the parent may only be brought in the case the subsidiary failed to indemnify its minority shareholders. See Art. 2497(3) c.c. Incidentally, requiring the subsidiary to pay the damages suffered by its non-controlling shareholders amounts to requiring plaintiff shareholders to pay themselves pro rata for the damage they suffered and are suing for. 160 See Enriques et al. (2017), at p 163 (similarly observing that 'successful challenge of an individual transaction harming a subsidiary will become more difficult [under group law]'). and the increased cost of capital for all firms. 161 To be sure, we cannot draw a firm conclusion that the EMCA regime on groups would be, all in all, inferior to using the common corporate law tools against self-dealing. But our analysis should be sufficient at least to cast doubt on the merits of special, more lenient rules on directors' liability such as under the EMCA.

Conclusion
Minority-co-owned groups are inherently problematic: on the one hand, they facilitate controllers' tunnelling and therefore minority shareholder expropriation; on the other, they may at times create value from the perspective not only of the controller but also for the other equity investors involved. 162 Director duties and more specific 'fair-dealing' rules governing decision-making processes over conflicted transactions are an important component of the basic toolkit traditionally used by corporate law to address minority-co-owned groups' heightened tunnelling risks. The efficacy of this component in curbing tunnelling is imperfect (for instance, non-transactional tunnelling techniques 163 are typically left unchecked). Furthermore, its very use generates transaction costs that inevitably hinder efficient intragroup exchange and, with that, firm (group) value maximisation.
In light of these shortcomings, a number of European legal scholars have argued that groups would be better regulated under a different regime based on the relaxation of ordinary corporate law barriers against unfair self-dealing. According to their view, directors of a group subsidiary should be allowed to take the broader group interest into account when making a decision. Accordingly, disadvantageous transactions that benefit the whole group should be permitted, provided that the subsidiary receives proper indemnification for the harm suffered, or even that such indemnification may, at the time the harmful decision is made, reasonably be expected to occur. Major EU jurisdictions, such as France and Italy, and to a lesser extent Germany, already provide groups with special rules consistent with this principle, and there is a chance that EU policymakers will follow suit, thus establishing a pan-European special group law in line with those scholars' proposals.
Our analysis of group law-as epitomised by the European Model Companies Act (EMCA) rules on corporate groups-has shown that the benefits of this special regime are limited. The EMCA regime for corporate groups has very limited enabling capacity: almost any type of group-value-maximising action that can be implemented under the EMCA principle (actions that require the sacrifice of the relevant affiliate's interest included) can also be implemented under ordinary corporate law. The efficiency contribution of the EMCA standard is that it allows some of those actions to be executed at a lower cost. However, the magnitude of this cost reduction appears to be modest. Furthermore, the EMCA standard does not appear to do a good job at providing for a more rounded assessment of fairness in intragroup exchange. The principle may help to reduce the number of false positives (i.e., transactions that do not divert value but are mistakenly considered to be value-diverting), but it correspondingly increases the number of false negatives, thereby increasing the risk of tunnelling, which in turn entails higher agency costs. As our analysis has shown, under the EMCA standard the risk of minority shareholder expropriation would in fact increase significantly, because such shareholders will find it more difficult to challenge controllers' value diversion and thus obtain indemnification for uncompensated harmful decisions.
Our conclusion leads us to suggest that, if special rules on groups such as those of the EMCA were to be adopted (a policy whose merits our analysis has cast significant doubts upon), they should be shaped as an opt-in regime, with the opt-in decision to be subject to stringent safeguards to prevent opportunistic opt-ins by companies that already feature minority shareholders in their shareholder base (such as publicly traded ones). An even better solution would be not to move European (Union) corporate law in the direction epitomised by the EMCA and for jurisdictions that have already espoused an EMCA-like solution to reconsider it.